This work proposes a feature-based technique to recognize vehicle typeswithin day and night times. Support vector machine (SVM) classifier is appliedon image histogram and CENsus Transformed histogRam Oriented Gradient (CENTROG)features in order to classify vehicle types during the day and night. Thermalimages were used for the night time experiments. Although thermal images sufferfrom low image resolution, lack of colour and poor texture information, theyoffer the advantage of being unaffected by high intensity light sources such asvehicle headlights which tend to render normal images unsuitable for night timeimage capturing and subsequent analysis. Since contour is useful in shape basedcategorisation and the most distinctive feature within thermal images, CENTROGis used to capture this feature information and is used within the experiments.The experimental results so obtained were compared with those obtained byemploying the CENsus TRansformed hISTogram (CENTRIST). Experimental resultsrevealed that CENTROG offers better recognition accuracies for both day andnight times vehicle types recognition.
展开▼